Neighbor Discovery in a Sensor Network with Directional Antennae

Abstract

Consider a network of n directional antennae in the plane. We consider the problem of efficient neighbor discovery in a (synchronous) network of sensors employing directional antennae. In this setting sensors send messages and listen for messages by directing their antennae towards a specific direction (which is not necessarily known in advance). In our model the directional antennae can be rotated by the sensors as required so as to discover all neighbors in their vicinity. In this paper we will limit ourselves to the (D,D) communication model whereby sensors employ directional antennae with identical transmission/reception beam widths. Our methodology is based on techniques for symmetry breaking so as to enable sender/receiver communication. We provide 1) deterministic algorithms that introduce delay in the rotation of the antennae and exploit knowledge of the existence of a vertex coloring of the network, and 2) randomized algorithms that require knowledge only of an upper bound on the size of the network so as to accomplish neighbor discovery. In both instances we study tradeoffs on the efficiency of the algorithms proposed.